What problem does it solve?
Handling Feishu email through an AI agent requires careful control over reading, drafting, and sending so that no message is sent prematurely and no sensitive content leaks. This Skill provides a safe, structured workflow for Feishu mailbox operations with draft-first sending and explicit confirmation gates.
Core Features & Use Cases
- Inbox Triage and Reading: List unread mail, search by query/folder/label, fetch single messages, batch messages, and full threads with sender context.
- Draft-First Sending:
mail send saves a draft by default; real sending only happens with explicit --confirm-send after verifying recipients, subject, and attachments.
- Replies, Forwards, and HTML Mail: Automatic Re:/Fwd: prefixes, quoted blocks, In-Reply-To threading, HTML body auto-detection, and CID inline image upload for rich emails.
- Use Case: Ask the agent to triage unread inbox messages, summarize an email thread, draft a weekly report email with inline charts, review the draft, and only send after your explicit confirmation.
Quick Start
Ask the agent to list your unread Feishu inbox messages and draft a reply to one of them without sending it.